Asymmetric alkylation of acyclic ketones via chiral metallo enamines

D. M. Brown and G. H. Jones, J. Chem. Soc. C, 252 (1967). (a) J. S. Brimacombe ... Roger W. Binkley,* David G. Hehemann. Department of Chemistry. Clev...
0 downloads 0 Views 355KB Size
J.Org. Chem., Vol. 43, No. 16, 1978

Communications

3245

28, 225 (1973); S. Hanessian, Adv. Chem. Ser., No. 74, 159 (1968); J. E.

G.Barnett, Adv. Carbohydr. Chem., 22, 177 (1967). M. Fieser and L. F. Fieser, "Reagents for Organic Synthesis", Vol. 4, Wiley, New York, N.Y., 1974, p 533. Pfanstiehl Laboratories, Waukegan, 111. (a) L. D. Hall and D.C. FAiller, Carbohydr. Res., 40, C1 (1975); (b) L. D. Hall and D.C. Miller, ibid., 47, 229 (1976). The procedwe for triflate formation used in the present study is a modification of that employed by Hall and Miller. Aldrich Chemical Company, inc., Milwaukee, Wisc. R. H. Bell, D. Horton, D. M. Williams, and E. Winter-Mihaly, Carbohydr. Res., 58, 109 (1977). D. M. Brown and G. H. Jones, J. Chem. Soc. C, 252 (1967). (a)J. S.Brimacombe. A 8. Foster, R. Hems, J. H. Westwood, and L. D. Hall, Can. J. Chem., 48, 3046 (1970); (b) A. E. Foster, R. Hems, and J. M. Webber, Carbohydr. Res., 5, 292 (1967). (a) S.Hanessian and N. R. Plessas, J. Org. Chem., 34,2163 (1969); (b) S. Hanessian. Methods Carbohydr. Chem., 6, 190 (1972). S.Hanessian, M. M Ponpipom, and P. Lavallee, Carbohydr. Res., 24,45

I

(RI-1

5

LDA, R I __9

HQAc-OAc-,

THF

R

Asymmetric Alkylation of Acyclic Ketones via Chiral Metallo Enamines. Effect of Kinetic vs. Thermodynamic Metalations S u m m a r y : Chiral imines of acyclic ketones have been metalated and alkylated t o afford a-alkyl ketones in 20-98% ee after equilibration of the metallo enamines. Sir: We recently described an asymmetric synthesis of 2alkylcyclohexanones (2) via the chiral methoxyamine 1.l The process was based upon the premise that the intermediate metallo enamine 3 would exist as a rigid five-membered ring through chelation of the methoxyl group with the lithium cation, providing favorable topological features for specific sites of alkylation. In order to account for the configurations

o$

/

+ (R)-l

'H

R" 7

6

C.R. Haylock, L. D. Melton. K. N. Slessor, and A. S.Tracey, Carbohydr.

Roger W. Binkley,* David G . Hehemann Department of Chemistry Cleveland S t a t e University Cleveland, Ohio 44115 Received March 24, 1978

OMe

A

118731 I -,.

Res., 16, 375 (1971). A. Zamojski, W. A. Szarek, and J. K. N. Jones, Carbohydr. Res., 26, 208 119731. D.~C.C. Smith, S. Chem. SOC.,1244(1956). E. Hardegger, G. Zanetti, and K. Steiner. Helv. Chim. Acta, 46, 282 (1963). K . James and S. J. Angyal. Aust. J. Chem., 25, 1967 (1972). H.M. Kissman and B. R. Baker, J. Am. Chem. SOC.,79, 5534 (1957).

I

CH2R'

OMe

tioselective alkylation, but a minor modification of the experimental procedure led to a-alkyl ketones 7 in generally high enantiomeric purity (Table I) within the constraints of certain structural features (vide infra). When the reaction conditions utilized for 2 (LDA, -20 O C ; R"1, -78 "C) were employed for the acyclic imines 5, alkylation occurred with no difficulty, furnishing 6 (Scheme I). However, hydrolysis to the ketones 7 gave products in 3-44% enantiomeric excess (Table I, % ee's in parentheses). The possibility that the lithiated enamines were kinetically formed as mixtures of E and 2 isomers (8A, SB), a situation not possible in the cyclohexene system 3, was therefore considered. When the lithio enamines, formed at -20 "C, were heated to reflux (THF) prior to alkylation, and the alkyl halide was added after cooling at -78 "C (first three entires, Table I), the ketones were obtained in 76-98% ee! In the case where R (in 4 or 5) was phenyl or benzyl (last four entries, Table I), the % ee of the ketones 7 was less dramatically affected. Thus, the kinetic or the thermodynamic ratio of lithio enamines 8 is dependent upon the bulk of the substituents present. It may, therefore, be assumed that metalation of 5 produces 8A and 8B kinetically; however, heating 8 (A, B) allows equilibration which favors the most stable lithio enamine. The relative size of the three substituents

I

OMe 5

I

OMe (R)-l 3

2 R =Me, Et, Pr, allyl, PhCH,

obtained, it was suggested that the alkyl halide may approach from the front side of the cyclohexene moiety, leading to 2alkylcyclohexanones in high enantiomeric purity.2 It seemed reasonable that this rigid metallo enamine should also provide a-alkyl ketones of high enantiomeric purity in the acyclic series. We now report that the conditions utilized above with acyclic ketones gave very poor results with regard to enan0022-3263/78/1943-3245$01.00/0

about the double bond will dictate whether the thermodynamic ratio of enamines is very different from the kinetic ratio.3 As seen from the table, the progression of high % ee to a relatively lower one follows the increasing size of substituents on the ketone or lithio enamine 8. Larger groups, such as phenyl, may decrease the thermodynamic stability of the E relative to the 2 isomer, resulting in ketones of lower enantiomeric purity. In the last entry of Table I, two phenyl groups are present in 8 and from the absolute configuration of the ketone isolated, we may assume that the (2)-enamine is slightly favored (60:40)to give 20% ee with an S configuration. The previous conditions for generating lithio enamines (--20 O C ) were obviously insufficient to effect their equilibration

0 1978 American Chemical Society

3246 J . Org. Chem., Vol. 43, No. 16, 1978 Table I. Formation of Chiral a-Alkyl Ketones (7)

-

R

Communications

a-alkyl ketones

4

R'

R"I

7

+

Yield, %

%eea

[aIz5D

configuration

Me

n-Bu

n-Pr

Me1

75

-12.8' (2.6,Et201

88 (435)

48'

$24.3" (2.9,Et20)

76 (3)

-15.3" (5.6,EtZO)

35

$23.5' (5.4,EtzO)

53 (44)

90s

-12.0" (4.3,PhH)

41 (25)

88s

f51.40 (1.5,EtOH)

20

0

n-Pr

Et

Me1

+ Me

Et

Me

Et1

Ph

Et

Me1

Ph

Me

Et1

0

Me

ph+ 0

Ph

6 Me

PhCHz

Ph

Me1

Ph+Ph 0

Ph

Ph

Me1

PhJ

Ph 0

a Values of % ee in parentheses are those obtained prior to heating 8 for 2 h. Percent enantiomeric excess determined by chiral purchased from Aldrich. Rotation based shift reagent, tris[3-(Xleptafluoro-l-hydroxybutylidene)-d-camphorato]europium(III), on 17.5" (1.8,ether) for pure ketone: D. Seebach,V. Ehrig and M. Teschner, Justus Liebigs Ann. Chem., 1357 (1976).Chiral shift reagent, as in b, confirmed enantiomeric purity. F. Nerdel and E. Henkel, Chem.Ber., 86,1002 (1953).e Low yield due to partial loss of material during solvent evaporation. f D. Seebach, D. Steinmuller, and F. Demuth, Angeu;, Chem., Znt. Ed. Engl., 7,620 (1968).g Contains unalkylated ketone, [(Y]D values are extrapolated after known mixtures were prepared to validate such extrapolations. The amount of unalkylated ketone in each case varied from 7 to 30% and the [(Y]D varied linearly. See ref 2a. Absolute configuration determined by Baeyer-Villiger oxidation of 7 to a-methylbenzyl alcohol of the R configuration. J Based on [ a ] D $252' (1.4,ethanol) reported by 0. Cervinka and L. Hub, Collect. Czech. Chem. Commun., 33,1911 (1968).

and they remain in their kinetic ratio until the solution is heated to reflux. An operationally useful mechanism for the alkylation, consistent with the absolute configuration of all the ketones produced, indicates that the alkyl iodide approaches the E isomer from the front side of the T system, analogous to 3, aligning both molecules enroute to the transition state. Although other mechanistic alternatives may be invoked,2bit is feasible to utilize the current scheme in order to predict the correct stereochemistry of the products. A typical procedure for (R)-(-)-3-methyl-4-heptanone follows. A mixture OF 10 g (60.6mmol) of 1: 8.3g (73 mmol) of 4-heptanone, and 100 mL of benzene was treated with 1-2 drops of trifluoroacetic acid and heated to reflux with azeotropic removal of water. The benzene solution was shaken with powdered sodium bicarbonate, dried, and distilled to give the imine 5 (R = n-Pr; R' = Et): bp 85-88 "C (0.02 Torr); 13.8g (85%); [aIz5~ +53.0" (4.76,MeOH). A solution of 21 mmol of lithium diisopropylamide in 40 mL of T H F was treated at -20 "C with 20 mmol of 5 dissolved in 20 mL of THF. The solution was stirred (-20 "C) for 1 h a n d then heated a t reflux for 2 h. After cooling to -78 "C, a solution of 21 mmol of methyl iodide in 20 mL of T H F was added over 5 min and the reaction mixture stirred at -I78 "C for 4 h. The mixture was quenched with 2 mL of methanol and partitioned between ether (75mL) and water (75 mL). The aqueous phase was extracted (2X) with ether and the combined organic extracts were washed successively with water and brine. The solvents were evaporated and the crude imine 6 was directly dissolved in 75 mL of pentane and shaken for 30 min with 80 mL of acetic acid-

sodium acetate buffer.6 Separation of the layers, concentration, and distillation gave (R)-(-)-3-methyl-4-heptanone (85%), [a]D -17.1' ( c 2.0,EtzO), 98% ee.7 Further studies to extend the scope of this asymmetric alkylation to other carbonyl compounds are under investigation.8

Acknowledgment. The authors express their gratitude to the National Science Foundation and the donors of the Petroleum Research Fund, administered by the American Chemical Society, for financial support. References and Notes (1) A. I. Meyers, D. R. Williams, and M. Druelinger, J. Am. Chem. Soc.,98,3032 (1976). (2) Additional successes involving asymmetric ketone alkylation have since been reported: (a) D. Enders and H. Eichenauer, Angew. Chem., ht. Ed. Engl., 15, 549 (1976); (b) J. K. Whitesell and M. A. Whitesell. J. Org. Chem., 42, 377 (1977). (3) The kinetic ratio of 8A to 88 appears to be sensitive (as it should be) to highly hindered bases. Thus, lithium 2,2,6,64etramethylpiperideat -20 ' C was treated with the imine of 3-pentanone 5 (R = Et, R' = Me) and without heating to reflux (nonequilibratingconditions) gave, afler alkylation with ethyl iodide, ethyl sec-butyl ketone (third entry in Table i) in 69% ee. (4) Preparation of (R)-(+)-l was accomplished by BH3.Me2S reduction5 of (Rephenylalanine (G. D. Searle) to (R)-phenylalaninol [88%, [ a ] D +24.5' (1.2, EtOH)] and treatment with KH-Me1 to 1 [75%, [ a ] D +14.4" (5.0, PhH)]. (5) C. F. Lane, US. Patent (to Aldrich) 3 935 280. (6) G. R. Kieczykowski,R. H. Schlessinger, and R. B. Salsky, TetrahedronLett., 597 (1976) have described a useful buffer solution for the hydrolysis of imines. (7) The sensitivity of these ketones to racemization has been determined and found to be completely stable to VPC (150 "C).the sodium acetate-acetic acid buffer solution used in the hydrolysis of 8 even afler 24 h contact, and storage at room temperature for at least 4 weeks. The racemization under

*

J. Org. Chem., Vol. 43, No. 16,1978 3247

Communications any of these conditions was